A neuroevolutionary reinforcement learning implementation on the game of pong.
Just open the index.html file in a browser, the training starts automatically. Press the left arrow button to switch between single paddle/entire population. Use the slider below the canvas to speed up the game and hence the learning process.
To load the learnt model (BestPaddleGen27), set the loadBestBrain variable to true in sketch.js.
Pong game based off of a coding challenge by Daniel Shiffman: https://github.com/CodingTrain/website/tree/master/CodingChallenges/CC_067_Pong/P5
Toy neural network library: https://github.com/CodingTrain/Toy-Neural-Network-JS